Intelligent textiles, variously know as smart fabrics, electronic textiles, or e-textiles, have attracted considerable attenti

Smart skins

Numerous miniaturized sensors and actuators have been fabricated since the emergence of MEMS technology. Since the majorities of MEMS fabrication processes are either directly borrowed or derived from IC technology. Inherently, most MEMS devices are built on rigid substrates such as silicon and glass wafers. On the other hand, for a wide variety of applications, it has long been desirable that sensors, actuators, or circuits can be fabricated on flexible substrates so as to be mounted on nonplanar surfaces or even on flexible objects such as a human body. One example is the tactile sensor, which need to be flexible in order to be attached to curved shapes like fingers and arms. In the field of fluid monitoring and controlling, it is often of great interest to know the profile of certain physical parameters such as pressure, or shear stress distributions on a nonplanar surface. To address the challenges arising from these applications, we need to develop a technology which can enable the fabrication of micromachined sensors on flexible substrate, namely, a MEMS skin technology. Additionally, integration of circuits is highly desirable since this integration promises to bring very important benefits such as operational improvement, packaging simplification, and cost reduction. Therefore, we need to develop a flexible skin technology which is compatible with both ICs and MEMS.

There are many methods to fabricate flexible miniaturized transducers. The most straightforward method to make flexible transducers/electronics is to fabricate directly on flexible substrates similar to the fabrication of thin film transistors on plastic/polymer/metal substrates. The advantages of this method are low cost and the ability to make large area flexible structures. Nevertheless, due to the temperature limit imposed by the flexible substrates, many high temperature processes are ruled out and the material properties are not optimized. Transducers that need high temperature process or use rigid materials such as single crystal silicon are difficult to fabricate on flexible substrate. Furthermore, electronics can not be integrated using the mainstream IC technology. Although circuits based on amorphous silicon/conductive polymers are active research topics now, they will not catch up with the complexity and performance of circuits based on single crystalline silicon in the foreseeable future. Moreover, when the substrate is subjected to bending, the devices on it undergo the stress as well. This may cause two undesirable consequences: 1) the devices on the flexible substrates may crack if the bending curvature is too large; 2) the performance of the devices is affected by the substrate bending.

Dr. Xu developed a unique silicon flexible skin technology that is totally different from the traditional method. The conceptual fabrication process is illustrated in the following figure. Assuming that MEMS devices or ICs have already been fabricated on the silicon substrate, the first step of the skin fabrication is to coat a polymer layer on the front of the wafer. Then the polymer layer is patterned to expose metal pads. Note that if necessary, MEMS and ICs can be exposed as well at this step. After this, the silicon wafer is thinned down and etched through from the back to form the arrays of silicon islands by Deep Reactive Ion Etching (DRIE). Finally, another layer of polymer is coated on the back to encapsulate the silicon islands.

The basic structure of the silicon flexible skin is arrays of silicon islands sandwiched between two layers of polymers. MEMS devices and ICs are on rigid islands. When the skin is bent, the devices on islands will not be subjected to stress. At the same time, the islands are small enough that the flexibility is not impaired. The most important advantage of this technology is its compatibility with current MEMS and IC technologies, since MEMS devices and ICs can be fabricated on the silicon wafer before the formation of the skin. Not only significant R&D efforts can be saved by avoiding re-invention, but also abundant sensing and computation capabilities offered by the silicon-based technology can be readily integrated.

Using this technology, Dr. Xu developed flexible shear-stress sensor skins for flow separation detection. The sensor skin contains a 1-D array of 36 shear-stress sensors, which can cover the 180° surface of the half-inch diameter semi-cylinder with 5° resolution (Fig. 1). These sensor skins have been installed on an Unmanned Aerial Vehicle (UAV) and flight-tested successfully (Fig. 2). Real-time flow distribution around the leading edge was obtained by the sensor skin.

Figure 1. The shear-stress sensor skin mounted on a 0.5” diameter aluminum block.

Figure 2. The unmanned aerial vehicle installed with the shear-stress sensor skin

The first IC-integrated flexible shear-stress sensor skin has also been demonstrated using a post-CMOS MEMS process. This sensor skin contains on-chip bias and signal conditioning circuitry. Therefore, the packaging is significantly simplified and the reliability is improved. Figure 3 shows a fabricated IC-integrated skin. The silicon islands and the metal traces across the islands can be clearly observed.

Figure 3. One IC-integrated skin held in tweezers

Flexible underwater shear-stress sensor skins have also be developed for underwater applications. The flexible skin structure also allows a novel packaging scheme depicted in Fig 4. The metal traces can be folded and get to the backside of the plug through the slit due to the flexibility of the device. Then we can perform wire bonding or soldering on the back side of the plug conveniently. By this method, the flow disturbance introduced is minimized and the reliability is improved by isolating the electrical contact from the water. Figure 5 shows a picture of a sensor skin packaged on an aluminum block. We can see the electrical wires coming from the backside of the plug. Both the acquisition of shear-stress distributions on non-planar surfaces and a reliable packaging scheme are made possible by the smart skin technology.

Figure 4. Packaging scheme based on flexible skin structure.

Figure 5. Sensor skin packaged on an aluminum plug

The majority of this work was done by Dr. Xu at the Micromachining Lab at the California Institute of Technology. At Wayne State University, Dr. Xu is exploring several new exciting applications of this smart skin technology. One specific example is intelligent textiles.

References:

Y. Xu, F. Jiang, Y.-C. Tai, A. Huang, C.-M. Ho, and S. Newbern, "Flexible Shear-Stress Sensor Skin and its Application to Unmanned Aerial Vehicle", Sensors and Actuators A: Physical, vol. 105, pp. 321-329, 2003.
Y. Xu, Y.-C. Tai, A. Huang, and C.-M. Ho, "IC-Integrated Flexible Shear-Stress Sensor Skin," Journal of Microelectromechanical Systems, vol. 12, pp. 740-747, 2003.
Y. Xu, J. Clendenen, S. Tung, F. Jiang, and Y.-C. Tai, "Underwater flexible shear-stress sensor skins," The 17th IEEE International Conference on Micro Electro Mechanical Systems (MEMS), Maastricht, The Netherlands, 2004.